Animals shifted from a preferred to a less preferred reward perform more poorly than animals that experienced only the less preferred reward. Evidence from previous experiments suggests that the response to reward reduction (negative contrast) involves a sequence of cognitive and emotional processes. The proposed experiments will be concerned with why some anxiolytics, such as the benzodiazepine tranquilizers and ethanol, are not effective in alleviating contrast until some period of time after the reward shift, whereas other anxiolytics, such as barbiturates, are effective immediately. Specific experiments will be concerned with: (a) the hypothesis that the neurotransmitter GABA is involved in recovery from contrast, but this recovery cycle requires time and/or experience with the postshift solution before it becomes activated; (b) the hypothesis that the initial reaction to reward loss involves a search process, that benzodiazepine tranquilizers will not be effective while the search process is ongoing, and that the failure of the search process activates a relevant GABAergic circuit involved in contrast recovery; (c) the hypothesis that degree of environmental complexity will modify the duration of contrast and the effectiveness of benzodiazepine tranquilizers in alleviating contrast. Other experiments will be directed toward uncovering the mechanism of action of cyproheptadine, a potent contrast antagonist; with the role of uncertainty following reward reduction in promoting elevations in corticosteroids; and with the relative effectiveness of novel tranquilizers whose actions are specific for different benzodiazepine receptor subtypes.